Research map / Reinforcement learning / Policy gradient and actor-critic
Security Analysis of Safe and Seldonian Reinforcement Learning Algorithms
Pinar Ozisik, Philip S. Thomas
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add Panacea clipping to IS safety test, alpha=0.5 | CH, IS safety test · Diabetes | adversarial trajectories needed: 49 -> 59 | Improved |
| add Panacea clipping to IS safety test, alpha=0.5 | CH, IS safety test · Grid-world | adversarial trajectories needed: 49 -> 68 | Improved |
| add Panacea clipping to WIS safety test, alpha=0.01 | CH, WIS safety test · Diabetes | adversarial trajectories needed: 1 -> 65 | Improved |
| Panacea with IS at alpha=0.1 | CH, IS safety test · Both domains | passes unsafe policy | Improved |
| Panacea with WIS on grid-world | CH, WIS safety test · Grid-world | passes unsafe policy | Improved |
| increase alpha from 0.5 to 5 in Panacea (IS) | Panacea · Diabetes | adversarial trajectories needed: 59 -> 49 | Hurt |
| increase alpha from 0.5 to 5 in Panacea (IS) | Panacea · Grid-world | adversarial trajectories needed: 68 -> 50 | Hurt |
| increase alpha from 0.5 to 1 in Panacea (WIS) | Panacea · Diabetes | adversarial trajectories needed: 65 -> 1 | Hurt |
| alpha larger than alpha' of existing methods | Panacea · Both domains | adversarial trajectories needed | No clear effect |
Closest papers by meaning
- Multi-level Certified Defense Against Poisoning Attacks in Offline Reinforcement Learning (ICLR 2025 · similarity 0.89)
- PolicyGuard: Towards Test-time and Step-level Adversary Defense for Reinforcement Learning Agent (ICML 2026 · similarity 0.87)
- Towards Robust Offline Reinforcement Learning under Diverse Data Corruption (ICLR 2024 · similarity 0.87)
- A Provable Approach for End-to-End Safe Reinforcement Learning (NeurIPS 2025 · similarity 0.87)
- Optimal Attack and Defense for Reinforcement Learning (AAAI 2024 · similarity 0.87)
- Safe Reinforcement Learning with Instantaneous Constraints: The Role of Aggressive Exploration (AAAI 2024 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
